Application of Multi level Deep Convolutional Network with Adaptive Space and Dynamic Upsampler in Elderly Micro expression Recognition
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
With aging, skin sags, wrinkles grow, and facial structures change, complicating microexpression recognition. Traditional technology may see accuracy decline. This study optimized microexpression recognition systems. By using ASF, YOLOv8, and a dynamic upsampler, the system's mAP rose from 0.951 to 0.97. It performs better in complex environments, especially with low-quality images and subtle expressions.
